DopplerNet: a convolutional neural network for recognising targets in real scenarios using a persistent range–Doppler radar

Roldan, Ignacio; del-Blanco, Carlos R.; Quevedo, Álvaro Duque de; Urzaiz, Fernando Ibañez; Menoyo, Javier Gismero; López, A.; Berjón, Daniel; Jaureguizar, Fernando; García, Narciso

doi:10.1049/iet-rsn.2019.0307

Cited by 55 publications

(33 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As seen from the results, the CNN classifier has 99.75% accuracy for the learned targets, with 100% distinction between humans and cars. This is consistent with the results of state-of-art research using the CNN classifier in Roldan et al [22]. However, the extremely high accuracy for the learned target may lead to an overconfidence problem that misclassifies unknown class target as one of the learned classes.…”

Section: Performance Evaluation Of Proposed Methodssupporting

confidence: 89%

See 1 more Smart Citation

Radar target classification considering unknown classes using deep convolutional neural network ensemble

Lee

Kang

et al. 2021

IET Radar Sonar & Navi

View full text Add to dashboard Cite

The target classification of unknown classes using radar sensor data is discussed. The neural network-based classifier shows high classification accuracy for the learned class targets. However, there is a risk of false decision for the untrained class target owing to an overconfidence problem. The output confidence of the classifier is calibrated using the deep convolutional neural network ensemble structure to propose a method to set the proper threshold for output confidence to decide unknown class targets. When using the proposed method, the accuracy of the learned target is maintained similar to that of the existing single neural network-based classifier, whereas the unknown class target is better identified. Further analysis verifies the effectiveness of the proposed method using commercial automotive radar. The proposed method can classify learned targets with an accuracy of 95% and distinguish unknown class targets with an accuracy of at least 85%. Based on the interaction with other sensors, individual sensors need to make reserved decisions about uncertain information. It is expected that the proposed ensemble network will be efficient in designing the classifier to perform target classification including unknown class decision.This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

show abstract

Section: Performance Evaluation Of Proposed Methodssupporting

confidence: 89%

“…As a benchmark, we compared the results with that of a single CNN-based classifier, which is a state-of-the-art method for classifying targets from similar radar data [22,23]. The configuration and hyperparameters of the CNN model were tuned to fit the input data and provide the best performance.…”

Section: Performance Evaluation Of Proposed Methodsmentioning

confidence: 99%

Radar target classification considering unknown classes using deep convolutional neural network ensemble

Lee

Kang

et al. 2021

IET Radar Sonar & Navi

View full text Add to dashboard Cite

show abstract

“…The authors of [7] applied principal component analysis (PCA) to the time-frequency distributions of both direct-path and multi-path signals to extract meaningful features of the small UAVs that are successively fused. Finally, in [8], the authors developed a framework to detect and classify drones in a 978-1-7281-8942-0/20/$31.00 ©2020 IEEE persistent range-Doppler radar. Following a CFAR detection stage, a convolutional neural network is used to perform the target recognition.…”

Section: Introductionmentioning

confidence: 99%

A Feature-Based Approach for Loaded/Unloaded Drones Classification Exploiting micro-Doppler Signatures

Pallotta

Clemente

Raddi

et al. 2020

2020 IEEE Radar Conference (RadarConf20)

View full text Add to dashboard Cite

“…Further, prior works [1,2,43] have been showing that mmW radars are with excellent environmental resistance 19 To compare the complexity between one 4D model and RAMP-CNN model model, we replace the 3D convolution kernels in RODNet-CDC model with the 4D convolution kernels and call the new model 4D-CDC. 20 Note: we didn't implement the 4D-CDC model, so the prediction time and layer numbers are ignored here. 21 The number of conv layers and transposed conv layers in models.…”

Section: Discussionmentioning

confidence: 99%

RAMP-CNN: A Novel Neural Network for Enhanced Automotive Radar Object Recognition

et al. 2021

View full text Add to dashboard Cite

Millimeter-wave (mmW) radars are being increasingly integrated into commercial vehicles to support new advanced driver-assistance systems (ADAS) by enabling robust and high-performance object detection, localization, as well as recognition-a key component of new environmental perception. In this paper, we propose a novel radar multiple-perspectives convolutional neural network (RAMP-CNN) that extracts location and class of objects based on further processing of the rangevelocity-angle (RVA) heatmap sequences. To bypass the complexity of 4D convolutional neural networks (NN), we propose to combine several lower-dimension NN models within our RAMP-CNN model that nonetheless approaches the performance upperbound with lower complexity. The extensive experiments show that the proposed RAMP-CNN model achieves better average recall (AR) and average precision (AP) than prior works in all testing scenarios (see Table.III). Besides, the RAMP-CNN model is validated to work robustly under the nighttime, which enables low-cost radars as a potential substitute for pure optical sensing under severe conditions.

show abstract

DopplerNet: a convolutional neural network for recognising targets in real scenarios using a persistent range–Doppler radar

Cited by 55 publications

References 17 publications

Radar target classification considering unknown classes using deep convolutional neural network ensemble

Radar target classification considering unknown classes using deep convolutional neural network ensemble

A Feature-Based Approach for Loaded/Unloaded Drones Classification Exploiting micro-Doppler Signatures

RAMP-CNN: A Novel Neural Network for Enhanced Automotive Radar Object Recognition

Contact Info

Product

Resources

About